HIV remains a global public health priority and occurs disproportionality among minority and stigmatized populations such as men who have sex with men (MSM). The occurrence of HIV is not random in the population, rather it exhibits patterns by individual and contextual factors. Is it the contextual factors, such as the neighborhood where one lives, socializes, or engages in sex, that is the focus of the proposed research. This proposed program of training and research examines the geospatial features of HIV using epidemiological data and analyses. The overarching goal is to improve the epidemiology of studying HIV by incorporating geospatial measures into HIV risk models through three areas. First, opportunities exist to tailor and refine existing HIV surveillance programs towards the most vulnerable groups using contextual determinants of risk. A second opportunity is to improve the validity of geospatial analyses through more accurate measurement and correct quantification of the spatial unit. A final opportunity is to improve risk quantification and reduce biased inference by applying spatial features of disease distribution to studying HIV risk. Taken together, attention to these three insights can improve HIV surveillance and epidemiology to guide prevention services. This project utilizes Philadelphia Department of Public Health?s AIDS Activities Coordinating Office surveillance data of people living with HIV/AIDS in Philadelphia in conjunction with data obtained via systematic review. Research questions to be answered include the optimal placement of HIV screening services in Philadelphia, the most appropriate choice of contextual unit to analyze spatial data (e.g., neighborhood, ZIP code, census tract), and the impact of sexual networks based upon geographic location of partner selection. Analyses will include advanced geospatial methods, hierarchical regression modeling, and agent-based simulation models. Follow-up work will include joint systems dynamic and agent-based models with a focus on health disparities in HIV. The proposed research will be complemented by experiential and didactic training in (1) geographic information systems and spatial analysis of epidemiologic data, (2) systems science and health disparities, and (3) epidemiological surveillance systems with a focus on HIV surveillance. This research and training will enable a program of independent research on the contextual factors relating neighborhood and HIV transmission and infection among vulnerable and stigmatized populations.